Now you can make realistic videos from photos, Byte Dance introduced the IA model
We have all managed a lot of Tiktok. But did you hear the name of Bytedance. In fact, Tiktok’s parent company, Bytedance, has now invented an AI model that has created panic worldwide. Let me tell you that the name of this new model is omnihuman-1. It is a multimodal AI technique, which converts any photo and audio into ultra-realistic video.
Deepfack concerns have increased
Due to the arrival of this new technology, there was a debate on Deepfek video. It is said that with the help of this model, any photo can be converted into video and that any audio can be added to it. Which will be very difficult to identify whether the video is real or modified. In the video made via Omnihuman-1, facial gestures and movements seem so natural that they seem to be like a real human being. This model only produces videos with photos, as well as speeches and songs can also be added.
How does it work now
According to Bytedance, this model uses data mixing technology, which combines with facial gestures and expressions as well as the whole movement of the body. Not only that, these facial expressions have movements and lip gestures in a very natural way, which makes the video ready-made very real. Society affirms that this model of AI can bring a revolutionary change in virtual entertainment and online education. Thanks to this, it is preferable for new possibilities in the fields of films, animations, virtual education and digital avatars.
